pytesseract识别验证码

Posted 2022-01-26 Harris-H

pytesseract识别验证码

0.下载

https://tesseract-ocr.github.io/tessdoc/Installation.html
`

1.安装

选择添加Math 和Chinese 包

安装完成后，配置环境变量，之前我安装1201版本的会报错，
win10 64位下会报这种错误：

所以我选择20190623的

安装后配置环境变量。

2.测试

识别结果

3.使用pytesseract

安装

pip install pytesseract

找到安装路径下的pytesseract.py

修改为自己OCR的路径

https://www.cnblogs.com/xiao-apple36/p/8865387.html#_label3_3

from pytesseract import image_to_string
from PIL import Image


def depoint(image):  # 像素 判断一个点周围情况 4，8邻域
    """
    降噪
    :param image:
    :return:
    """
    pixdata = image.load()
    print(pixdata)
    w, h = image.size
    for y in range(1, h - 1):
        for x in range(1, w - 1):
            count = 0
            if pixdata[x, y - 1] > 245:
                count += 1
            if pixdata[x, y + 1] > 245:
                count += 1
            if pixdata[x - 1, y] > 245:
                count += 1
            if pixdata[x + 1, y] > 245:
                count += 1
            if count > 3:
                pixdata[x, y] = 255
    return image


def binaring(image, threshold=160):
    """
    对传入的图像进行灰度，二值化处理
    :param image:
    :param threshold:
    :return:
    """
    image = image.convert('L')
    image.show()
    pixdata = image.load()
    # print(pixdata)
    w, h = image.size
    for y in range(h):
        for x in range(w):
            # print(pixdata[x,y])
            if pixdata[x, y] < threshold:
                pixdata[x, y] = 0
            else:
                pixdata[x, y] = 255
    return image


if __name__ == '__main__':
    # image = Image.open('img.png')
    # # pix_l = []
    # # image.show()
    # # pix_l_set = sorted(list(set(pix_l)))
    # # print(pix_l_set[:len(pix_l_set)//2])  # 求平均数的值
    # image2 = binaring(image)  # 二值化
    # image3 = depoint(image2)  # 降噪
    # image3.show()
    # # 识别文字
    # print('code: ', image_to_string(image3, lang='eng'))
    image = Image.open('test.png')
    print(image_to_string(image, lang='chi_sim'))

https://www.cnblogs.com/xiao-apple36/p/8747351.html